preprocess
Jianing Yao
2023-03-01
Step 1: read data and clean
# read data
# the spreadsheet has 19 metabolic measures and 5 retinal thickness
data <- as.data.frame(read_excel("/Users/jianingyao/Desktop/School/JHU/Year2_Term3/Practice_Statistical_Consulting/Project/statistical_consulting_retina_project/Data/Sachdeva_Final_datasheet_OCT_metabolic_232023.xlsx"))# convert id and gender to factors
data$IRB <- as.factor(data$IRB)
data$gender <- as.factor(data$gender)
levels(data$gender) <- c("male", "female")
data <- mutate(data, group = ifelse(HbA1c > 6.4, "DM", ifelse(HbA1c < 5.7, "Control", "Prediabetes")))
data$group <- as.factor(data$group)Step 2: descriptive statistics
# gender
table(data$gender)##
## male female
## 18 22table(data$gender)/nrow(data)##
## male female
## 0.45 0.55# diabetes groups
table(data$group)##
## Control DM Prediabetes
## 16 12 12table(data$group)/nrow(data)##
## Control DM Prediabetes
## 0.4 0.3 0.3table 1: mean and sd of metabolic measures and retinal measures before imputation
cols_ind <- c(2, 4:112)
min_data <- apply(data[, cols_ind], 2, function(x) min(x, na.rm = TRUE))
max_data <- apply(data[, cols_ind], 2, function(x) max(x, na.rm = TRUE))
median_data <- apply(data[, cols_ind], 2, function(x) median(x, na.rm = TRUE))
mean_data <- apply(data[, cols_ind], 2, function(x) mean(x, na.rm = TRUE))
sd_data <- apply(data[, cols_ind], 2, function(x) sd(x, na.rm = TRUE))
n_data <- apply(data[, cols_ind], 2, function(x) sum(!is.na(x)))
summary <- cbind(min_data, max_data, median_data, mean_data, sd_data, n_data)
colnames(summary) <- c("Min", "Max", "Median", "Mean", "SD", "N")
round(summary,2)## Min Max Median Mean SD N
## age 57.00 81.00 71.50 71.40 6.18 40
## heart_rate 50.00 100.00 69.00 70.44 10.86 39
## MAP 58.00 127.00 95.00 94.91 12.82 37
## SBP 105.00 187.00 145.50 146.40 22.48 40
## BMI 20.80 45.20 27.95 29.70 6.33 38
## HbA1c 4.90 10.00 5.85 6.29 1.27 40
## BUN 9.00 31.00 15.00 16.77 5.13 40
## Cr 0.50 1.90 1.00 1.01 0.27 40
## BUN/Cr 9.00 26.00 16.00 17.05 4.15 40
## HDL 35.00 118.00 55.00 59.54 19.47 39
## LDL 37.00 143.00 86.00 88.16 29.12 37
## Triglycerides 98.19 265.12 136.43 146.02 42.06 39
## Cholesterol 144.03 301.71 206.83 207.35 35.94 39
## NEFA 0.45 1.71 0.98 0.99 0.25 39
## Glucose 75.85 181.71 90.49 99.58 25.62 39
## Ketone 0.02 1.47 0.09 0.15 0.23 39
## Insulin 0.97 128.85 10.21 19.12 25.36 36
## Adiponectin 1.09 32.20 8.41 11.31 8.95 39
## Leptin 1.51 65.00 22.20 25.21 16.31 39
## RAGE 3.89 39.13 12.10 12.64 5.94 39
## CSF_TOT_OD 234.00 382.00 272.00 274.51 26.89 37
## CSF_NFL_OD 8.00 18.00 13.00 12.35 2.42 37
## CSF_(GCL-IPL)_OD 23.00 65.00 33.00 36.70 8.93 37
## CSF_INL_OD 12.00 41.00 21.00 21.81 6.45 37
## CSF_ORT_OD 46.00 124.00 68.00 70.86 16.28 37
## SI_TOT_OD 296.00 370.00 329.00 333.38 16.22 37
## SI_NFL_OD 18.00 37.00 23.00 23.62 3.23 37
## SI_(GCL-IPL)_OD 72.00 113.00 89.00 88.97 8.71 37
## SI_INL_OD 33.00 52.00 40.00 41.22 4.36 37
## SI_ORT_OD 95.00 131.00 115.00 114.65 8.11 37
## SO_TOT_OD 266.00 313.00 286.00 289.03 13.63 37
## SO_NFL_OD 30.00 54.00 37.00 37.49 5.02 37
## SO_(GCL-IPL)_OD 49.00 70.00 58.00 58.41 5.63 37
## SO_INL_OD 27.00 36.00 31.00 30.68 2.19 37
## SO_ORT_OD 79.00 111.00 97.00 97.27 7.69 37
## NI_TOT_OD 298.00 380.00 334.00 336.09 17.88 35
## NI_NFL_OD 16.00 36.00 20.00 20.83 3.36 35
## NI_(GCL-IPL)_OD 73.00 117.00 89.00 88.54 9.15 35
## NI_INL_OD 33.00 52.00 41.00 41.17 4.93 35
## NI_ORT_OD 102.00 135.00 122.00 119.83 8.89 35
## NO_TOT_OD 268.00 345.00 301.00 304.20 15.97 35
## NO_NFL_OD 35.00 66.00 47.00 47.94 6.31 35
## NO_(GCL-IPL)_OD 50.00 79.00 62.00 62.03 6.89 35
## NO_INL_OD 29.00 38.00 33.00 32.89 2.32 35
## NO_ORT_OD 78.00 116.00 98.00 96.83 8.98 35
## II_TOT_OD 291.00 379.00 330.00 331.97 17.98 36
## II_NFL_OD 19.00 35.00 23.50 23.83 3.37 36
## II_(GCL-IPL)_OD 70.00 111.00 87.00 88.58 9.36 36
## II_INL_OD 34.00 49.00 40.50 40.72 4.27 36
## II_ORT_OD 96.00 127.00 115.00 114.06 8.72 36
## IO_TOT_OD 252.00 308.00 279.00 281.42 15.55 36
## IO_NFL_OD 18.00 52.00 37.50 37.81 6.88 36
## IO_(GCL-IPL)_OD 48.00 72.00 56.50 58.25 6.38 36
## IO_INL_OD 26.00 37.00 31.00 30.72 2.43 36
## IO_ORT_OD 75.00 108.00 91.50 91.50 8.15 36
## TI_TOT_OD 284.00 380.00 321.00 322.86 18.73 37
## TI_NFL_OD 11.00 23.00 17.00 16.92 1.96 37
## TI_(GCL-IPL)_OD 66.00 110.00 84.00 84.81 10.26 37
## TI_INL_OD 31.00 49.00 38.00 38.46 4.27 37
## TI_ORT_OD 97.00 133.00 117.00 117.30 7.80 37
## TO_TOT_OD 253.00 307.00 272.00 275.11 14.43 37
## TO_NFL_OD 15.00 26.00 18.00 18.70 1.97 37
## TO_(GCL-IPL)_OD 51.00 74.00 64.00 63.78 6.00 37
## TO_INL_OD 28.00 38.00 32.00 32.41 2.55 37
## TO_ORT_OD 75.00 113.00 95.00 95.41 7.87 37
## CSF_TOT_OS 231.00 356.00 276.00 276.53 25.19 38
## CSF_NFL_OS 8.00 35.00 13.00 13.39 4.06 38
## CSF_(GCL-IPL)_OS 23.00 91.00 36.50 37.66 11.84 38
## CSF_INL_OS 10.00 37.00 22.00 22.29 6.18 38
## CSF_ORT_OS 113.00 166.00 131.50 134.47 11.50 38
## SI_TOT_OS 301.00 350.00 331.00 330.65 12.48 37
## SI_NFL_OS 19.00 32.00 23.00 23.46 3.05 37
## SI_(GCL-IPL)_OS 75.00 98.00 86.00 86.19 7.11 37
## SI_INL_OS 33.00 55.00 41.00 41.24 4.41 37
## SI_ORT_OS 92.00 136.00 114.00 114.59 9.33 37
## SO_TOT_OS 270.00 340.00 285.00 288.84 14.18 37
## SO_NFL_OS 29.00 53.00 36.00 37.78 5.71 37
## SO_(GCL-IPL)_OS 48.00 76.00 57.00 57.89 5.73 37
## SO_INL_OS 25.00 37.00 30.00 30.62 2.56 37
## SO_ORT_OS 77.00 120.00 97.00 97.76 8.24 37
## NI_TOT_OS 302.00 354.00 333.50 332.92 13.07 36
## NI_NFL_OS 16.00 48.00 20.00 21.17 5.16 36
## NI_(GCL-IPL)_OS 71.00 99.00 86.50 86.72 8.32 36
## NI_INL_OS 34.00 55.00 40.50 41.06 4.80 36
## NI_ORT_OS 104.00 140.00 117.50 118.36 9.26 36
## NO_TOS_OS 267.00 339.00 298.00 299.19 13.82 36
## NO_NFL_OS 35.00 84.00 47.00 49.00 9.10 36
## NO_GCL-IPL)_OS 49.00 71.00 59.50 58.86 5.33 36
## NO_INL_OS 27.00 45.00 33.00 32.64 3.24 36
## NO_ORT_OS 0.00 128.00 94.00 91.76 18.25 37
## II_TOT_OS 293.00 353.00 327.00 327.16 13.54 38
## II_NFL_OS 18.00 48.00 23.00 24.13 4.89 38
## II_(GCL-IPL)_OS 68.00 98.00 84.00 84.16 7.81 38
## II_INL_OS 31.00 51.00 40.00 40.92 4.53 38
## II_ORT_OS 94.00 135.00 113.00 113.45 10.10 38
## IO_TOT_OS 250.00 309.00 279.00 279.63 13.58 38
## IO_NFL_OS 29.00 63.00 38.00 38.32 6.73 38
## IO_(GCL-IPL)_OS 45.00 72.00 54.50 56.42 5.58 38
## IO_INL_OS 25.00 35.00 30.50 30.37 2.25 38
## IO_ORT_OS 74.00 203.00 90.50 93.87 20.08 38
## TI_TOT_OS 289.00 345.00 320.00 320.05 13.79 37
## TI_NFL_OS 15.00 27.00 17.00 17.81 2.15 37
## TI_(GCL-IPL)_OS 67.00 94.00 79.00 80.76 8.00 37
## TI_INL_OS 28.00 45.00 37.00 36.76 3.97 37
## TI_ORT_OS 101.00 150.00 116.00 118.43 9.35 37
## TO_TOT_OS 252.00 317.00 274.00 278.22 15.03 37
## TO_NFL_OS 17.00 29.00 20.00 20.05 2.13 37
## TO_(GCL-IPL)_OS 54.00 78.00 62.00 63.57 6.07 37
## TO_INL_OS 28.00 37.00 32.00 32.14 2.18 37
## TO_OR_OS 75.00 116.00 95.00 97.05 8.52 37summary <- round(summary,2)
write.csv(summary, "summary_beforeImp.csv", row.names = TRUE, col.names = TRUE)## Warning in write.csv(summary, "summary_beforeImp.csv", row.names = TRUE, :
## attempt to set 'col.names' ignoredsummary by groups before imputation
cols_ind <- c(4:113)
summary_group <- data[cols_ind] %>%
group_by(group) %>%
summarize_all(list(median = ~ median(., na.rm = TRUE),
mean = ~ mean(., na.rm = TRUE),
sd = ~ sd(., na.rm = TRUE)))
as.data.frame(t(summary_group))## V1 V2 V3
## group Control DM Prediabetes
## heart_rate_median 68.0 77.5 66.0
## MAP_median 95 95 100
## SBP_median 150 139 152
## BMI_median 27.15 29.80 25.70
## HbA1c_median 5.40 7.55 6.00
## BUN_median 14.0 16.5 14.5
## Cr_median 0.95 1.00 0.95
## BUN/Cr_median 15.5 19.0 16.0
## HDL_median 60.5 52.0 52.0
## LDL_median 93.5 63.5 93.0
## Triglycerides_median 128.6820 137.9845 140.8270
## Cholesterol_median 224.5730 191.1265 208.8735
## NEFA_median 0.894 1.000 0.997
## Glucose_median 86.3410 108.1710 90.8535
## Ketone_median 0.0630 0.0945 0.0845
## Insulin_median 12.718 16.922 9.219
## Adiponectin_median 8.4130 7.2545 9.0810
## Leptin_median 18.479 24.794 21.736
## RAGE_median 11.3010 13.0220 11.2445
## CSF_TOT_OD_median 279.0 273.5 263.5
## CSF_NFL_OD_median 14.0 12.5 12.0
## CSF_(GCL-IPL)_OD_median 35.0 36.5 33.0
## CSF_INL_OD_median 21.0 24.5 17.0
## CSF_ORT_OD_median 68.0 75.5 64.5
## SI_TOT_OD_median 329 336 326
## SI_NFL_OD_median 24.0 23.5 23.0
## SI_(GCL-IPL)_OD_median 89.0 89.5 88.5
## SI_INL_OD_median 41 40 40
## SI_ORT_OD_median 117 116 114
## SO_TOT_OD_median 286.0 287.0 281.5
## SO_NFL_OD_median 38 36 37
## SO_(GCL-IPL)_OD_median 57.0 58.0 59.5
## SO_INL_OD_median 31.0 30.5 30.5
## SO_ORT_OD_median 98.0 98.5 93.5
## NI_TOT_OD_median 335 340 324
## NI_NFL_OD_median 21.5 20.0 20.0
## NI_(GCL-IPL)_OD_median 90.5 89.5 87.0
## NI_INL_OD_median 41.5 40.5 40.0
## NI_ORT_OD_median 123.5 122.5 116.0
## NO_TOT_OD_median 303 303 299
## NO_NFL_OD_median 49.5 46.5 45.0
## NO_(GCL-IPL)_OD_median 57.5 63.0 62.0
## NO_INL_OD_median 34.0 32.5 33.0
## NO_ORT_OD_median 99.0 97.5 91.0
## II_TOT_OD_median 333.5 333.0 323.5
## II_NFL_OD_median 23.5 23.5 23.5
## II_(GCL-IPL)_OD_median 86 87 88
## II_INL_OD_median 42 39 40
## II_ORT_OD_median 117.0 118.0 112.5
## IO_TOT_OD_median 276.5 281.0 277.5
## IO_NFL_OD_median 36 38 37
## IO_(GCL-IPL)_OD_median 57.0 58.5 56.0
## IO_INL_OD_median 31.5 30.0 30.5
## IO_ORT_OD_median 93.0 91.5 89.0
## TI_TOT_OD_median 324.0 324.5 313.0
## TI_NFL_OD_median 16.0 17.0 16.5
## TI_(GCL-IPL)_OD_median 86.0 84.5 81.5
## TI_INL_OD_median 38.0 38.0 36.5
## TI_ORT_OD_median 119.0 115.5 115.5
## TO_TOT_OD_median 274 275 269
## TO_NFL_OD_median 18 19 18
## TO_(GCL-IPL)_OD_median 63 65 63
## TO_INL_OD_median 32.0 32.5 32.0
## TO_ORT_OD_median 97.0 94.5 92.0
## CSF_TOT_OS_median 278.5 289.5 259.5
## CSF_NFL_OS_median 13 13 12
## CSF_(GCL-IPL)_OS_median 37.5 38.0 34.0
## CSF_INL_OS_median 21.5 25.5 21.0
## CSF_ORT_OS_median 136.0 133.5 131.0
## SI_TOT_OS_median 325.0 340.0 325.5
## SI_NFL_OS_median 23.5 22.0 23.0
## SI_(GCL-IPL)_OS_median 85.5 86.0 87.0
## SI_INL_OS_median 42.0 41.0 40.5
## SI_ORT_OS_median 115.5 117.0 113.0
## SO_TOT_OS_median 283.0 291.0 285.5
## SO_NFL_OS_median 35.5 36.0 37.0
## SO_(GCL-IPL)_OS_median 56.0 61.0 57.5
## SO_INL_OS_median 31 30 29
## SO_ORT_OS_median 96.5 99.0 96.5
## NI_TOT_OS_median 330.5 339.0 326.0
## NI_NFL_OS_median 20 20 19
## NI_(GCL-IPL)_OS_median 88 88 85
## NI_INL_OS_median 42.5 40.0 40.0
## NI_ORT_OS_median 122 118 114
## NO_TOS_OS_median 296 300 297
## NO_NFL_OS_median 48.5 52.0 46.0
## NO_GCL-IPL)_OS_median 58 60 60
## NO_INL_OS_median 33 33 32
## NO_ORT_OS_median 95 94 89
## II_TOT_OS_median 323.5 329.5 326.0
## II_NFL_OS_median 23.0 23.5 23.0
## II_(GCL-IPL)_OS_median 84.0 83.0 86.5
## II_INL_OS_median 41.0 40.5 40.0
## II_ORT_OS_median 114.5 114.5 110.0
## IO_TOT_OS_median 279 279 279
## IO_NFL_OS_median 39.0 37.5 37.5
## IO_(GCL-IPL)_OS_median 54.5 54.5 55.5
## IO_INL_OS_median 31.0 30.5 30.0
## IO_ORT_OS_median 90.5 89.5 90.5
## TI_TOT_OS_median 314.5 324.0 320.0
## TI_NFL_OS_median 17 17 17
## TI_(GCL-IPL)_OS_median 80.0 79.0 77.5
## TI_INL_OS_median 36.5 37.0 37.0
## TI_ORT_OS_median 116.5 120.0 114.5
## TO_TOT_OS_median 272.5 284.0 274.0
## TO_NFL_OS_median 20.0 20.0 19.5
## TO_(GCL-IPL)_OS_median 61.5 64.0 61.0
## TO_INL_OS_median 33.0 33.0 30.5
## TO_OR_OS_median 95.5 100.0 93.5
## heart_rate_mean 65.68750 78.50000 68.54545
## MAP_mean 97.31250 90.40909 96.00000
## SBP_mean 151.6250 137.3333 148.5000
## BMI_mean 29.66250 31.62545 27.81818
## HbA1c_mean 5.306250 7.891667 5.991667
## BUN_mean 15.3125 18.5000 17.0000
## Cr_mean 0.9937500 0.9833333 1.0500000
## BUN/Cr_mean 15.81250 19.08333 16.66667
## HDL_mean 66.06250 55.50000 54.45455
## LDL_mean 99.37500 64.80000 93.09091
## Triglycerides_mean 142.1453 148.8373 148.0620
## Cholesterol_mean 226.0751 186.9147 204.3800
## NEFA_mean 0.9403333 1.0457500 0.9808333
## Glucose_mean 92.39027 117.94717 90.20325
## Ketone_mean 0.22693333 0.11316667 0.08341667
## Insulin_mean 13.14214 26.89909 18.96400
## Adiponectin_mean 11.78700 10.02042 12.01775
## Leptin_mean 27.14433 24.11900 23.88725
## RAGE_mean 11.48153 15.73908 10.99292
## CSF_TOT_OD_mean 283.2667 273.8333 262.2000
## CSF_NFL_OD_mean 13.26667 11.83333 11.60000
## CSF_(GCL-IPL)_OD_mean 40.20 35.25 33.20
## CSF_INL_OD_mean 22.40 22.75 19.80
## CSF_ORT_OD_mean 75.86667 69.83333 64.60000
## SI_TOT_OD_mean 334.4667 334.0833 330.9000
## SI_NFL_OD_mean 24.26667 23.41667 22.90000
## SI_(GCL-IPL)_OD_mean 88.06667 89.50000 89.70000
## SI_INL_OD_mean 41.46667 41.41667 40.60000
## SI_ORT_OD_mean 115.1333 114.8333 113.7000
## SO_TOT_OD_mean 289.4667 289.7500 287.5000
## SO_NFL_OD_mean 38.80000 36.33333 36.90000
## SO_(GCL-IPL)_OD_mean 57.13333 59.41667 59.10000
## SO_INL_OD_mean 30.8 30.5 30.7
## SO_ORT_OD_mean 97.46667 97.66667 96.50000
## NI_TOT_OD_mean 340.2143 337.4167 327.8889
## NI_NFL_OD_mean 21.92857 20.41667 19.66667
## NI_(GCL-IPL)_OD_mean 89.42857 88.25000 87.55556
## NI_INL_OD_mean 41.42857 41.58333 40.22222
## NI_ORT_OD_mean 120.9286 121.0000 116.5556
## NO_TOT_OD_mean 306.5714 305.0000 299.4444
## NO_NFL_OD_mean 49.57143 46.75000 47.00000
## NO_(GCL-IPL)_OD_mean 61.42857 62.58333 62.22222
## NO_INL_OD_mean 33.57143 32.66667 32.11111
## NO_ORT_OD_mean 97.28571 97.91667 94.66667
## II_TOT_OD_mean 335.2857 332.4167 326.8000
## II_NFL_OD_mean 24.5 24.0 22.7
## II_(GCL-IPL)_OD_mean 89.07143 88.08333 88.50000
## II_INL_OD_mean 41.78571 39.91667 40.20000
## II_ORT_OD_mean 114.6429 115.5000 111.5000
## IO_TOT_OD_mean 283.4286 282.0833 277.8000
## IO_NFL_OD_mean 38.28571 38.08333 36.80000
## IO_(GCL-IPL)_OD_mean 58.92857 58.41667 57.10000
## IO_INL_OD_mean 31.14286 30.41667 30.50000
## IO_ORT_OD_mean 92.71429 91.25000 90.10000
## TI_TOT_OD_mean 326.6667 323.8333 316.0000
## TI_NFL_OD_mean 16.73333 17.25000 16.80000
## TI_(GCL-IPL)_OD_mean 86.66667 84.66667 82.20000
## TI_INL_OD_mean 38.66667 39.08333 37.40000
## TI_ORT_OD_mean 118.6667 117.0833 115.5000
## TO_TOT_OD_mean 275.4667 276.2500 273.2000
## TO_NFL_OD_mean 18.80000 18.91667 18.30000
## TO_(GCL-IPL)_OD_mean 63.20000 64.58333 63.70000
## TO_INL_OD_mean 32.53333 32.25000 32.40000
## TO_ORT_OD_mean 96.26667 95.33333 94.20000
## CSF_TOT_OS_mean 278.4286 286.0000 264.8333
## CSF_NFL_OS_mean 13.28571 14.50000 12.41667
## CSF_(GCL-IPL)_OS_mean 38.64286 40.66667 33.50000
## CSF_INL_OS_mean 21.78571 25.25000 19.91667
## CSF_ORT_OS_mean 135.2857 136.0000 132.0000
## SI_TOT_OS_mean 330.1429 335.8182 326.5000
## SI_NFL_OS_mean 23.85714 23.54545 22.91667
## SI_(GCL-IPL)_OS_mean 85.78571 86.63636 86.25000
## SI_INL_OS_mean 41.71429 42.00000 40.00000
## SI_ORT_OS_mean 113.9286 117.4545 112.7500
## SO_TOT_OS_mean 286.1429 295.2727 286.0833
## SO_NFL_OS_mean 37.92857 38.18182 37.25000
## SO_(GCL-IPL)_OS_mean 55.85714 60.36364 58.00000
## SO_INL_OS_mean 30.71429 31.18182 30.00000
## SO_ORT_OS_mean 97.07143 100.09091 96.41667
## NI_TOT_OS_mean 334.4286 336.9091 327.0000
## NI_NFL_OS_mean 20.78571 23.45455 19.36364
## NI_(GCL-IPL)_OS_mean 87.21429 86.00000 86.81818
## NI_INL_OS_mean 41.85714 41.90909 39.18182
## NI_ORT_OS_mean 118.7857 119.3636 116.8182
## NO_TOS_OS_mean 298.7857 304.1818 294.7273
## NO_NFL_OS_mean 48.50000 50.72727 47.90909
## NO_GCL-IPL)_OS_mean 58.21429 59.45455 59.09091
## NO_INL_OS_mean 33.07143 32.90909 31.81818
## NO_ORT_OS_mean 88.26667 96.18182 92.09091
## II_TOT_OS_mean 327.1429 329.9167 324.4167
## II_NFL_OS_mean 23.78571 25.16667 23.50000
## II_(GCL-IPL)_OS_mean 84.50000 82.08333 85.83333
## II_INL_OS_mean 41.85714 41.50000 39.25000
## II_ORT_OS_mean 112.2857 115.7500 112.5000
## IO_TOT_OS_mean 279.3571 282.5000 277.0833
## IO_NFL_OS_mean 37.92857 39.08333 38.00000
## IO_(GCL-IPL)_OS_mean 56.28571 56.50000 56.50000
## IO_INL_OS_mean 31.14286 30.00000 29.83333
## IO_ORT_OS_mean 90.35714 92.16667 99.66667
## TI_TOT_OS_mean 319.8571 323.1818 317.4167
## TI_NFL_OS_mean 17.64286 18.45455 17.41667
## TI_(GCL-IPL)_OS_mean 81.42857 80.00000 80.66667
## TI_INL_OS_mean 37.14286 37.90909 35.25000
## TI_ORT_OS_mean 117.7143 120.1818 117.6667
## TO_TOT_OS_mean 275.5714 283.4545 276.5000
## TO_NFL_OS_mean 19.71429 20.72727 19.83333
## TO_(GCL-IPL)_OS_mean 62.28571 65.63636 63.16667
## TO_INL_OS_mean 32.50000 32.63636 31.25000
## TO_OR_OS_mean 96.50000 99.09091 95.83333
## heart_rate_sd 7.905431 12.551711 7.916611
## MAP_sd 11.694550 16.466357 9.591663
## SBP_sd 25.42669 18.99442 20.37155
## BMI_sd 7.602182 6.166127 3.949891
## HbA1c_sd 0.2594064 1.1277156 0.2234373
## BUN_sd 4.045059 5.213619 6.105139
## Cr_sd 0.2264766 0.2480225 0.3450955
## BUN/Cr_sd 3.816084 2.998737 5.033223
## HDL_sd 22.35910 15.44197 17.68255
## LDL_sd 29.01235 16.54489 27.75051
## Triglycerides_sd 45.68817 43.13420 39.51485
## Cholesterol_sd 41.57926 23.37620 27.78883
## NEFA_sd 0.2284696 0.3132745 0.2202968
## Glucose_sd 25.44691 28.32414 10.32736
## Ketone_sd 0.36366183 0.06555474 0.04030894
## Insulin_sd 9.110366 26.348271 36.499134
## Adiponectin_sd 8.883801 9.339039 9.280763
## Leptin_sd 22.59918 12.33036 10.42255
## RAGE_sd 3.148879 8.984779 3.663867
## CSF_TOT_OD_sd 35.28550 20.82757 11.38029
## CSF_NFL_OD_sd 2.491892 2.037527 2.503331
## CSF_(GCL-IPL)_OD_sd 11.213512 7.072675 5.050853
## CSF_INL_OD_sd 7.038669 5.545268 6.746192
## CSF_ORT_OD_sd 19.89568 13.62373 11.41344
## SI_TOT_OD_sd 16.47018 16.77367 16.62963
## SI_NFL_OD_sd 4.096456 3.028901 1.728840
## SI_(GCL-IPL)_OD_sd 9.391993 9.298094 7.645623
## SI_INL_OD_sd 4.257207 4.679905 4.526465
## SI_ORT_OD_sd 9.687007 7.234178 7.211873
## SO_TOT_OD_sd 12.74400 14.52975 15.13825
## SO_NFL_OD_sd 6.710333 3.725425 2.960856
## SO_(GCL-IPL)_OD_sd 5.938574 5.664215 5.321863
## SO_INL_OD_sd 2.007130 2.022600 2.790858
## SO_ORT_OD_sd 7.558029 7.749878 8.553752
## NI_TOT_OD_sd 20.50208 15.28789 15.67996
## NI_NFL_OD_sd 4.598734 1.928652 2.179449
## NI_(GCL-IPL)_OD_sd 10.725342 8.400487 8.323327
## NI_INL_OD_sd 4.449966 4.944388 6.016182
## NI_ORT_OD_sd 10.484159 8.168676 7.001984
## NO_TOT_OD_sd 18.19431 15.36229 13.60249
## NO_NFL_OD_sd 7.470087 5.940998 4.690416
## NO_(GCL-IPL)_OD_sd 8.563916 6.501165 4.841946
## NO_INL_OD_sd 2.680823 2.015095 2.027588
## NO_ORT_OD_sd 8.818063 9.238834 9.578622
## II_TOT_OD_sd 20.05542 17.46403 16.01943
## II_NFL_OD_sd 4.127767 3.190896 2.213594
## II_(GCL-IPL)_OD_sd 11.323660 8.596599 8.045012
## II_INL_OD_sd 4.117104 4.907477 3.765339
## II_ORT_OD_sd 9.660444 7.716570 8.822320
## IO_TOT_OD_sd 14.88786 15.18647 17.79388
## IO_NFL_OD_sd 7.268780 5.264950 8.508819
## IO_(GCL-IPL)_OD_sd 6.955510 6.359793 6.045200
## IO_INL_OD_sd 2.070197 2.274696 3.171050
## IO_ORT_OD_sd 7.770047 8.225515 9.170605
## TI_TOT_OD_sd 21.25245 16.92139 16.52607
## TI_NFL_OD_sd 2.463060 1.912875 1.135292
## TI_(GCL-IPL)_OD_sd 11.555250 9.354953 9.647107
## TI_INL_OD_sd 3.976119 5.053502 3.893014
## TI_ORT_OD_sd 9.393513 5.915439 7.531416
## TO_TOT_OD_sd 12.33964 14.14294 18.58195
## TO_NFL_OD_sd 2.336053 1.831955 1.636392
## TO_(GCL-IPL)_OD_sd 5.334524 6.200562 7.165504
## TO_INL_OD_sd 2.503331 2.301185 3.134042
## TO_ORT_OD_sd 8.163216 7.075352 8.941787
## CSF_TOT_OS_sd 21.06296 30.51378 20.68303
## CSF_NFL_OS_sd 2.163636 6.612660 1.975225
## CSF_(GCL-IPL)_OS_sd 7.611804 17.849540 7.403930
## CSF_INL_OS_sd 4.964157 7.059809 5.775471
## CSF_ORT_OS_sd 10.31376 11.39378 13.40285
## SI_TOT_OS_sd 12.15197 14.29558 10.16679
## SI_NFL_OS_sd 3.109715 3.531675 2.678478
## SI_(GCL-IPL)_OS_sd 7.234305 8.151966 6.552238
## SI_INL_OS_sd 4.268463 5.744563 3.104249
## SI_ORT_OS_sd 10.049602 9.842394 8.080560
## SO_TOT_OS_sd 10.067901 20.533343 9.802211
## SO_NFL_OS_sd 7.415828 5.437245 3.768892
## SO_(GCL-IPL)_OS_sd 4.817550 7.159228 4.690416
## SO_INL_OS_sd 2.812843 2.821992 2.044949
## SO_ORT_OS_sd 8.222523 10.133652 6.402533
## NI_TOT_OS_sd 14.11336 13.16400 10.30534
## NI_NFL_OS_sd 2.006856 8.430464 2.730301
## NI_(GCL-IPL)_OS_sd 7.637746 10.334409 7.652688
## NI_INL_OS_sd 4.896736 6.057302 2.713602
## NI_ORT_OS_sd 8.980137 9.982712 9.568889
## NO_TOS_OS_sd 13.486053 18.015650 7.603827
## NO_NFL_OS_sd 7.046549 13.557956 6.057302
## NO_GCL-IPL)_OS_sd 4.995052 6.919012 4.253341
## NO_INL_OS_sd 2.615465 4.657350 2.272364
## NO_ORT_OS_sd 25.808267 13.570690 6.564367
## II_TOT_OS_sd 14.65951 14.93902 11.09839
## II_NFL_OS_sd 3.641187 7.673251 2.067058
## II_(GCL-IPL)_OS_sd 7.572724 9.307459 6.562058
## II_INL_OS_sd 4.452435 5.728716 2.895922
## II_ORT_OS_sd 10.373254 9.696532 10.638694
## IO_TOT_OS_sd 12.38862 16.54471 12.19880
## IO_NFL_OS_sd 6.509925 9.159777 4.177864
## IO_(GCL-IPL)_OS_sd 4.842577 7.452882 4.641708
## IO_INL_OS_sd 1.561909 2.796101 2.249579
## IO_ORT_OS_sd 7.550926 10.861386 33.399329
## TI_TOT_OS_sd 13.69367 14.81093 13.54762
## TI_NFL_OS_sd 1.6458406 3.3574882 0.9962049
## TI_(GCL-IPL)_OS_sd 8.083357 8.921883 7.655460
## TI_INL_OS_sd 3.899845 4.085451 3.792936
## TI_ORT_OS_sd 8.712754 6.968761 12.160542
## TO_TOT_OS_sd 12.53128 17.55200 15.30597
## TO_NFL_OS_sd 1.540658 3.258555 1.337116
## TO_(GCL-IPL)_OS_sd 5.209881 7.352180 5.718126
## TO_INL_OS_sd 2.312175 2.062655 2.005674
## TO_OR_OS_sd 8.492078 8.848215 8.674239write.csv(t(summary_group), "summary_group_beforeImp.csv", row.names = TRUE, col.names = TRUE)## Warning in write.csv(t(summary_group), "summary_group_beforeImp.csv", row.names
## = TRUE, : attempt to set 'col.names' ignoredreplace missing values in metabolic measures with median
met_list <- c("heart_rate","MAP","SBP","BMI","HbA1c","BUN","Cr","BUN/Cr","HDL","LDL","Triglycerides","Cholesterol","NEFA","Glucose","Ketone","Insulin","Adiponectin","Leptin","RAGE")
data.median <- data
for (col in met_list) {
data.median[is.na(data.median[[col]]), col] <- median(data.median[[col]], na.rm = TRUE)
}replace missing values in metabolic and retinal measures by knn
# Perform KNN imputation with default settings
df.imp <- kNN(data)
df.imp <- df.imp[1:113]table 1: mean and sd of metabolic measures and retinal measures after imputation
cols_ind <- c(2, 4:112)
min_data <- apply(df.imp[, cols_ind], 2, function(x) min(x, na.rm = TRUE))
max_data <- apply(df.imp[, cols_ind], 2, function(x) max(x, na.rm = TRUE))
median_data <- apply(df.imp[, cols_ind], 2, function(x) median(x, na.rm = TRUE))
mean_data <- apply(df.imp[, cols_ind], 2, function(x) mean(x, na.rm = TRUE))
sd_data <- apply(df.imp[, cols_ind], 2, function(x) sd(x, na.rm = TRUE))
n_data <- apply(df.imp[, cols_ind], 2, function(x) sum(!is.na(x)))
summary <- cbind(min_data, max_data, median_data, mean_data, sd_data, n_data)
colnames(summary) <- c("Min", "Max", "Median", "Mean", "SD", "N")
round(summary,2)## Min Max Median Mean SD N
## age 57.00 81.00 71.50 71.40 6.18 40
## heart_rate 50.00 100.00 69.00 70.58 10.75 40
## MAP 58.00 127.00 95.50 95.09 12.35 40
## SBP 105.00 187.00 145.50 146.40 22.48 40
## BMI 20.80 45.20 27.95 29.66 6.21 40
## HbA1c 4.90 10.00 5.85 6.29 1.27 40
## BUN 9.00 31.00 15.00 16.77 5.13 40
## Cr 0.50 1.90 1.00 1.01 0.27 40
## BUN/Cr 9.00 26.00 16.00 17.05 4.15 40
## HDL 35.00 118.00 55.00 59.42 19.24 40
## LDL 37.00 143.00 86.00 89.12 28.54 40
## Triglycerides 98.19 265.12 137.21 145.95 41.52 40
## Cholesterol 144.03 301.71 207.51 207.46 35.48 40
## NEFA 0.45 1.71 0.99 0.99 0.25 40
## Glucose 75.85 181.71 90.61 99.55 25.29 40
## Ketone 0.02 1.47 0.09 0.15 0.23 40
## Insulin 0.97 128.85 9.71 18.26 24.17 40
## Adiponectin 1.09 32.20 8.26 11.19 8.86 40
## Leptin 1.51 65.00 23.12 25.44 16.17 40
## RAGE 3.89 39.13 12.01 12.59 5.87 40
## CSF_TOT_OD 234.00 382.00 271.00 274.10 25.88 40
## CSF_NFL_OD 8.00 18.00 13.00 12.40 2.33 40
## CSF_(GCL-IPL)_OD 23.00 65.00 33.00 36.42 8.64 40
## CSF_INL_OD 12.00 41.00 21.50 21.75 6.21 40
## CSF_ORT_OD 46.00 124.00 68.00 70.75 15.67 40
## SI_TOT_OD 296.00 370.00 332.00 334.68 16.28 40
## SI_NFL_OD 18.00 37.00 23.00 23.50 3.13 40
## SI_(GCL-IPL)_OD 72.00 113.00 89.00 88.92 8.54 40
## SI_INL_OD 33.00 52.00 40.50 41.20 4.19 40
## SI_ORT_OD 95.00 131.00 116.00 114.97 7.88 40
## SO_TOT_OD 266.00 313.00 287.00 290.40 14.01 40
## SO_NFL_OD 30.00 54.00 37.00 37.38 4.86 40
## SO_(GCL-IPL)_OD 49.00 70.00 57.50 58.40 5.57 40
## SO_INL_OD 27.00 36.00 31.00 30.82 2.17 40
## SO_ORT_OD 79.00 111.00 98.50 97.97 7.94 40
## NI_TOT_OD 298.00 380.00 335.50 336.88 17.32 40
## NI_NFL_OD 16.00 36.00 20.00 20.77 3.15 40
## NI_(GCL-IPL)_OD 73.00 117.00 89.00 88.47 8.69 40
## NI_INL_OD 33.00 52.00 41.00 41.20 4.61 40
## NI_ORT_OD 102.00 135.00 122.00 120.58 8.57 40
## NO_TOT_OD 268.00 345.00 301.00 304.92 15.93 40
## NO_NFL_OD 35.00 66.00 45.00 47.48 6.02 40
## NO_(GCL-IPL)_OD 50.00 79.00 62.00 62.02 6.59 40
## NO_INL_OD 29.00 38.00 33.00 33.08 2.26 40
## NO_ORT_OD 78.00 116.00 98.00 97.65 8.84 40
## II_TOT_OD 291.00 379.00 330.00 331.90 17.18 40
## II_NFL_OD 19.00 35.00 23.00 23.68 3.23 40
## II_(GCL-IPL)_OD 70.00 111.00 86.00 87.80 9.21 40
## II_INL_OD 34.00 49.00 41.00 40.83 4.08 40
## II_ORT_OD 96.00 127.00 116.50 114.60 8.49 40
## IO_TOT_OD 252.00 308.00 279.00 280.95 14.87 40
## IO_NFL_OD 18.00 52.00 36.50 37.35 6.68 40
## IO_(GCL-IPL)_OD 48.00 72.00 56.00 57.90 6.14 40
## IO_INL_OD 26.00 37.00 31.00 30.88 2.37 40
## IO_ORT_OD 75.00 108.00 92.00 91.75 7.81 40
## TI_TOT_OD 284.00 380.00 321.00 322.78 18.07 40
## TI_NFL_OD 11.00 23.00 17.00 16.98 1.90 40
## TI_(GCL-IPL)_OD 66.00 110.00 83.00 84.10 10.22 40
## TI_INL_OD 31.00 49.00 38.00 38.48 4.10 40
## TI_ORT_OD 97.00 133.00 118.50 117.55 7.58 40
## TO_TOT_OD 253.00 307.00 272.00 274.95 13.91 40
## TO_NFL_OD 15.00 26.00 18.00 18.65 1.92 40
## TO_(GCL-IPL)_OD 51.00 74.00 63.00 63.58 5.83 40
## TO_INL_OD 28.00 38.00 32.00 32.42 2.47 40
## TO_ORT_OD 75.00 113.00 96.00 95.62 7.61 40
## CSF_TOT_OS 231.00 356.00 277.00 277.10 24.79 40
## CSF_NFL_OS 8.00 35.00 13.00 13.50 3.99 40
## CSF_(GCL-IPL)_OS 23.00 91.00 36.50 37.50 11.58 40
## CSF_INL_OS 10.00 37.00 22.00 22.55 6.13 40
## CSF_ORT_OS 113.00 166.00 133.50 134.85 11.33 40
## SI_TOT_OS 301.00 350.00 331.00 331.88 12.76 40
## SI_NFL_OS 19.00 32.00 23.00 23.50 2.94 40
## SI_(GCL-IPL)_OS 75.00 98.00 87.00 86.47 6.91 40
## SI_INL_OS 33.00 55.00 41.00 41.27 4.24 40
## SI_ORT_OS 92.00 136.00 115.00 114.97 9.07 40
## SO_TOT_OS 270.00 340.00 286.50 290.25 14.53 40
## SO_NFL_OS 29.00 53.00 36.00 37.58 5.53 40
## SO_(GCL-IPL)_OS 48.00 76.00 57.50 58.35 5.75 40
## SO_INL_OS 25.00 37.00 31.00 30.88 2.62 40
## SO_ORT_OS 77.00 120.00 98.00 98.30 8.15 40
## NI_TOT_OS 302.00 354.00 335.00 334.60 13.42 40
## NI_NFL_OS 16.00 48.00 20.00 21.05 4.90 40
## NI_(GCL-IPL)_OS 71.00 99.00 88.50 87.33 8.11 40
## NI_INL_OS 34.00 55.00 41.00 41.33 4.74 40
## NI_ORT_OS 104.00 140.00 119.50 119.12 9.07 40
## NO_TOS_OS 267.00 339.00 299.50 301.22 14.47 40
## NO_NFL_OS 35.00 84.00 50.50 49.30 8.67 40
## NO_GCL-IPL)_OS 49.00 71.00 60.00 59.23 5.24 40
## NO_INL_OS 27.00 45.00 33.00 32.90 3.19 40
## NO_ORT_OS 0.00 128.00 95.00 92.55 17.76 40
## II_TOT_OS 293.00 353.00 327.50 328.20 13.97 40
## II_NFL_OS 18.00 48.00 23.00 24.10 4.80 40
## II_(GCL-IPL)_OS 68.00 98.00 85.00 84.20 7.61 40
## II_INL_OS 31.00 51.00 40.50 41.05 4.49 40
## II_ORT_OS 94.00 135.00 113.50 113.95 10.11 40
## IO_TOT_OS 250.00 309.00 279.00 280.80 14.19 40
## IO_NFL_OS 29.00 63.00 38.00 38.27 6.64 40
## IO_(GCL-IPL)_OS 45.00 72.00 55.00 56.85 5.76 40
## IO_INL_OS 25.00 35.00 31.00 30.45 2.22 40
## IO_ORT_OS 74.00 203.00 91.00 94.17 19.63 40
## TI_TOT_OS 289.00 345.00 321.50 321.20 13.86 40
## TI_NFL_OS 15.00 27.00 17.00 17.82 2.06 40
## TI_(GCL-IPL)_OS 67.00 94.00 80.00 81.00 7.73 40
## TI_INL_OS 28.00 45.00 37.00 36.77 3.81 40
## TI_ORT_OS 101.00 150.00 118.00 118.67 9.03 40
## TO_TOT_OS 252.00 317.00 274.50 279.48 15.14 40
## TO_NFL_OS 17.00 29.00 20.00 20.08 2.07 40
## TO_(GCL-IPL)_OS 54.00 78.00 63.00 64.10 6.20 40
## TO_INL_OS 28.00 37.00 33.00 32.33 2.20 40
## TO_OR_OS 75.00 116.00 96.50 97.50 8.34 40summary <- round(summary,2)
write.csv(summary, "summary_afterImp.csv", row.names = TRUE, col.names = TRUE)## Warning in write.csv(summary, "summary_afterImp.csv", row.names = TRUE, :
## attempt to set 'col.names' ignoredsummary by groups after imputation
cols_ind <- c(4:113)
summary_group <- df.imp[cols_ind] %>%
group_by(group) %>%
summarize_all(list(median = ~ median(., na.rm = TRUE),
mean = ~ mean(., na.rm = TRUE),
sd = ~ sd(., na.rm = TRUE)))
as.data.frame(t(summary_group))## V1 V2 V3
## group Control DM Prediabetes
## heart_rate_median 68.0 77.5 66.0
## MAP_median 95.0 95.0 98.5
## SBP_median 150 139 152
## BMI_median 27.15 31.05 25.70
## HbA1c_median 5.40 7.55 6.00
## BUN_median 14.0 16.5 14.5
## Cr_median 0.95 1.00 0.95
## BUN/Cr_median 15.5 19.0 16.0
## HDL_median 60.5 52.0 53.0
## LDL_median 93.5 69.0 92.5
## Triglycerides_median 131.2660 137.9845 140.8270
## Cholesterol_median 224.2320 191.1265 208.8735
## NEFA_median 0.9265 1.0000 0.9970
## Glucose_median 87.6825 108.1710 90.8535
## Ketone_median 0.1160 0.0945 0.0845
## Insulin_median 12.7180 14.0715 9.2450
## Adiponectin_median 7.9805 7.2545 9.0810
## Leptin_median 18.6295 24.7940 21.7360
## RAGE_median 11.0145 13.0220 11.2445
## CSF_TOT_OD_median 276.5 273.5 266.0
## CSF_NFL_OD_median 13.5 12.5 13.0
## CSF_(GCL-IPL)_OD_median 34.0 36.5 33.0
## CSF_INL_OD_median 21.0 24.5 18.5
## CSF_ORT_OD_median 67.5 75.5 67.5
## SI_TOT_OD_median 330.5 336.0 330.0
## SI_NFL_OD_median 23.5 23.5 22.5
## SI_(GCL-IPL)_OD_median 87.5 89.5 88.5
## SI_INL_OD_median 41.0 40.0 40.5
## SI_ORT_OD_median 117.0 116.0 114.5
## SO_TOT_OD_median 288.5 287.0 288.0
## SO_NFL_OD_median 37.5 36.0 36.5
## SO_(GCL-IPL)_OD_median 56.0 58.0 59.5
## SO_INL_OD_median 31.0 30.5 31.0
## SO_ORT_OD_median 98.5 98.5 96.5
## NI_TOT_OD_median 335 340 332
## NI_NFL_OD_median 20.5 20.0 20.5
## NI_(GCL-IPL)_OD_median 89.0 89.5 89.5
## NI_INL_OD_median 42.0 40.5 41.0
## NI_ORT_OD_median 123.5 122.5 119.5
## NO_TOT_OD_median 300 303 301
## NO_NFL_OD_median 48.0 46.5 45.0
## NO_(GCL-IPL)_OD_median 57.5 63.0 62.0
## NO_INL_OD_median 34.0 32.5 33.0
## NO_ORT_OD_median 99.0 97.5 98.0
## II_TOT_OD_median 330 333 327
## II_NFL_OD_median 23.0 23.5 23.0
## II_(GCL-IPL)_OD_median 85 87 86
## II_INL_OD_median 42.0 39.0 40.5
## II_ORT_OD_median 117 118 114
## IO_TOT_OD_median 275.5 281.0 281.0
## IO_NFL_OD_median 34.5 38.0 35.5
## IO_(GCL-IPL)_OD_median 55.0 58.5 56.0
## IO_INL_OD_median 32 30 31
## IO_ORT_OD_median 92.0 91.5 90.5
## TI_TOT_OD_median 321.0 324.5 317.5
## TI_NFL_OD_median 16 17 17
## TI_(GCL-IPL)_OD_median 85.0 84.5 80.5
## TI_INL_OD_median 38.5 38.0 37.5
## TI_ORT_OD_median 119.0 115.5 117.5
## TO_TOT_OD_median 272.5 275.0 271.0
## TO_NFL_OD_median 18 19 18
## TO_(GCL-IPL)_OD_median 62.5 65.0 62.5
## TO_INL_OD_median 32.0 32.5 32.0
## TO_ORT_OD_median 96.5 94.5 93.5
## CSF_TOT_OS_median 278.5 289.5 259.5
## CSF_NFL_OS_median 13.5 13.0 12.0
## CSF_(GCL-IPL)_OS_median 37.5 38.0 34.0
## CSF_INL_OS_median 22.0 25.5 21.0
## CSF_ORT_OS_median 138.5 133.5 131.0
## SI_TOT_OS_median 328.0 340.5 325.5
## SI_NFL_OS_median 24.0 22.5 23.0
## SI_(GCL-IPL)_OS_median 87 88 87
## SI_INL_OS_median 42.0 41.5 40.5
## SI_ORT_OS_median 117.0 117.5 113.0
## SO_TOT_OS_median 284.0 292.0 285.5
## SO_NFL_OS_median 35 36 37
## SO_(GCL-IPL)_OS_median 56.5 61.0 57.5
## SO_INL_OS_median 31 31 29
## SO_ORT_OS_median 97.5 99.0 96.5
## NI_TOT_OS_median 332.5 339.0 328.5
## NI_NFL_OS_median 20 20 19
## NI_(GCL-IPL)_OS_median 89.5 89.0 85.5
## NI_INL_OS_median 43 41 40
## NI_ORT_OS_median 124.0 118.5 114.5
## NO_TOS_OS_median 299.5 301.5 297.5
## NO_NFL_OS_median 51.0 52.0 46.5
## NO_GCL-IPL)_OS_median 59 60 60
## NO_INL_OS_median 33.5 33.0 32.5
## NO_ORT_OS_median 95.0 94.5 92.0
## II_TOT_OS_median 325.0 329.5 326.0
## II_NFL_OS_median 23.0 23.5 23.0
## II_(GCL-IPL)_OS_median 85.0 83.0 86.5
## II_INL_OS_median 41.0 40.5 40.0
## II_ORT_OS_median 118.0 114.5 110.0
## IO_TOT_OS_median 279 279 279
## IO_NFL_OS_median 39.0 37.5 37.5
## IO_(GCL-IPL)_OS_median 55.0 54.5 55.5
## IO_INL_OS_median 31.5 30.5 30.0
## IO_ORT_OS_median 91.0 89.5 90.5
## TI_TOT_OS_median 318.0 325.5 320.0
## TI_NFL_OS_median 17.0 17.5 17.0
## TI_(GCL-IPL)_OS_median 82.0 80.5 77.5
## TI_INL_OS_median 37 37 37
## TI_ORT_OS_median 118.5 120.5 114.5
## TO_TOT_OS_median 273.5 284.5 274.0
## TO_NFL_OS_median 20.0 20.0 19.5
## TO_(GCL-IPL)_OS_median 62.5 65.0 61.0
## TO_INL_OS_median 33.0 33.0 30.5
## TO_OR_OS_median 97.0 100.0 93.5
## heart_rate_mean 65.68750 78.50000 69.16667
## MAP_mean 97.31250 91.20833 96.00000
## SBP_mean 151.6250 137.3333 148.5000
## BMI_mean 29.66250 31.68167 27.64167
## HbA1c_mean 5.306250 7.891667 5.991667
## BUN_mean 15.3125 18.5000 17.0000
## Cr_mean 0.9937500 0.9833333 1.0500000
## BUN/Cr_mean 15.81250 19.08333 16.66667
## HDL_mean 66.0625 55.5000 54.5000
## LDL_mean 99.37500 72.33333 92.25000
## Triglycerides_mean 142.2082 148.8373 148.0620
## Cholesterol_mean 225.1706 186.9147 204.3800
## NEFA_mean 0.9436875 1.0457500 0.9808333
## Glucose_mean 92.75919 117.94717 90.20325
## Ketone_mean 0.22331250 0.11316667 0.08341667
## Insulin_mean 12.93069 25.44358 18.19192
## Adiponectin_mean 11.45044 10.02042 12.01775
## Leptin_mean 27.60000 24.11900 23.88725
## RAGE_mean 11.43444 15.73908 10.99292
## CSF_TOT_OD_mean 282.5000 273.8333 263.1667
## CSF_NFL_OD_mean 13.25000 11.83333 11.83333
## CSF_(GCL-IPL)_OD_mean 39.75000 35.25000 33.16667
## CSF_INL_OD_mean 22.37500 22.75000 19.91667
## CSF_ORT_OD_mean 75.31250 69.83333 65.58333
## SI_TOT_OD_mean 335.3125 334.0833 334.4167
## SI_NFL_OD_mean 24.12500 23.41667 22.75000
## SI_(GCL-IPL)_OD_mean 87.81250 89.50000 89.83333
## SI_INL_OD_mean 41.43750 41.41667 40.66667
## SI_ORT_OD_mean 115.3750 114.8333 114.5833
## SO_TOT_OD_mean 290.4375 289.7500 291.0000
## SO_NFL_OD_mean 38.68750 36.33333 36.66667
## SO_(GCL-IPL)_OD_mean 57.00000 59.41667 59.25000
## SO_INL_OD_mean 30.87500 30.50000 31.08333
## SO_ORT_OD_mean 97.56250 97.66667 98.83333
## NI_TOT_OD_mean 339.1875 337.4167 333.2500
## NI_NFL_OD_mean 21.68750 20.41667 19.91667
## NI_(GCL-IPL)_OD_mean 88.8125 88.2500 88.2500
## NI_INL_OD_mean 41.50000 41.58333 40.41667
## NI_ORT_OD_mean 121.3125 121.0000 119.1667
## NO_TOT_OD_mean 305.2500 305.0000 304.4167
## NO_NFL_OD_mean 48.87500 46.75000 46.33333
## NO_(GCL-IPL)_OD_mean 61.06250 62.58333 62.75000
## NO_INL_OD_mean 33.62500 32.66667 32.75000
## NO_ORT_OD_mean 97.50000 97.91667 97.58333
## II_TOT_OD_mean 334.1875 332.4167 328.3333
## II_NFL_OD_mean 24.18750 24.00000 22.66667
## II_(GCL-IPL)_OD_mean 87.87500 88.08333 87.41667
## II_INL_OD_mean 41.75000 39.91667 40.50000
## II_ORT_OD_mean 115.0000 115.5000 113.1667
## IO_TOT_OD_mean 282.0625 282.0833 278.3333
## IO_NFL_OD_mean 37.50000 38.08333 36.41667
## IO_(GCL-IPL)_OD_mean 58.37500 58.41667 56.75000
## IO_INL_OD_mean 31.31250 30.41667 30.75000
## IO_ORT_OD_mean 92.56250 91.25000 91.16667
## TI_TOT_OD_mean 326.1250 323.8333 317.2500
## TI_NFL_OD_mean 16.81250 17.25000 16.91667
## TI_(GCL-IPL)_OD_mean 85.81250 84.66667 81.25000
## TI_INL_OD_mean 38.68750 39.08333 37.58333
## TI_ORT_OD_mean 118.6875 117.0833 116.5000
## TO_TOT_OD_mean 275.1250 276.2500 273.4167
## TO_NFL_OD_mean 18.75000 18.91667 18.25000
## TO_(GCL-IPL)_OD_mean 62.93750 64.58333 63.41667
## TO_INL_OD_mean 32.62500 32.25000 32.33333
## TO_ORT_OD_mean 96.25000 95.33333 95.08333
## CSF_TOT_OS_mean 279.6250 286.0000 264.8333
## CSF_NFL_OS_mean 13.56250 14.50000 12.41667
## CSF_(GCL-IPL)_OS_mean 38.12500 40.66667 33.50000
## CSF_INL_OS_mean 22.50000 25.25000 19.91667
## CSF_ORT_OS_mean 136.125 136.000 132.000
## SI_TOT_OS_mean 332.25 336.75 326.50
## SI_NFL_OS_mean 23.87500 23.58333 22.91667
## SI_(GCL-IPL)_OS_mean 86.31250 86.91667 86.25000
## SI_INL_OS_mean 41.6875 42.0000 40.0000
## SI_ORT_OS_mean 114.6875 117.5833 112.7500
## SO_TOT_OS_mean 288.7500 296.4167 286.0833
## SO_NFL_OS_mean 37.56250 37.91667 37.25000
## SO_(GCL-IPL)_OS_mean 56.8125 60.7500 58.0000
## SO_INL_OS_mean 31.12500 31.41667 30.00000
## SO_ORT_OS_mean 98.06250 100.50000 96.41667
## NI_TOT_OS_mean 336.6250 337.9167 328.5833
## NI_NFL_OS_mean 20.68750 23.16667 19.41667
## NI_(GCL-IPL)_OS_mean 87.93750 86.33333 87.50000
## NI_INL_OS_mean 42.375 42.000 39.250
## NI_ORT_OS_mean 119.6875 119.9167 117.5833
## NO_TOS_OS_mean 301.3750 305.4167 296.8333
## NO_NFL_OS_mean 48.93750 50.83333 48.25000
## NO_GCL-IPL)_OS_mean 58.81250 59.41667 59.58333
## NO_INL_OS_mean 33.43750 33.08333 32.00000
## NO_ORT_OS_mean 89.12500 96.66667 93.00000
## II_TOT_OS_mean 329.7500 329.9167 324.4167
## II_NFL_OS_mean 23.75000 25.16667 23.50000
## II_(GCL-IPL)_OS_mean 84.56250 82.08333 85.83333
## II_INL_OS_mean 42.0625 41.5000 39.2500
## II_ORT_OS_mean 113.6875 115.7500 112.5000
## IO_TOT_OS_mean 282.3125 282.5000 277.0833
## IO_NFL_OS_mean 37.87500 39.08333 38.00000
## IO_(GCL-IPL)_OS_mean 57.375 56.500 56.500
## IO_INL_OS_mean 31.25000 30.00000 29.83333
## IO_ORT_OS_mean 91.56250 92.16667 99.66667
## TI_TOT_OS_mean 321.8125 324.1667 317.4167
## TI_NFL_OS_mean 17.68750 18.41667 17.41667
## TI_(GCL-IPL)_OS_mean 81.75000 80.33333 80.66667
## TI_INL_OS_mean 37.12500 37.83333 35.25000
## TI_ORT_OS_mean 118.1875 120.3333 117.6667
## TO_TOT_OS_mean 277.8750 284.5833 276.5000
## TO_NFL_OS_mean 19.75000 20.75000 19.83333
## TO_(GCL-IPL)_OS_mean 63.18750 66.25000 63.16667
## TO_INL_OS_mean 32.8125 32.7500 31.2500
## TO_OR_OS_mean 97.31250 99.41667 95.83333
## heart_rate_sd 7.905431 12.551711 7.848953
## MAP_sd 11.694550 15.942308 8.686458
## SBP_sd 25.42669 18.99442 20.37155
## BMI_sd 7.602182 5.882396 3.815389
## HbA1c_sd 0.2594064 1.1277156 0.2234373
## BUN_sd 4.045059 5.213619 6.105139
## Cr_sd 0.2264766 0.2480225 0.3450955
## BUN/Cr_sd 3.816084 2.998737 5.033223
## HDL_sd 22.35910 15.44197 16.86039
## LDL_sd 29.01235 23.65792 26.61894
## Triglycerides_sd 44.13969 43.13420 39.51485
## Cholesterol_sd 40.33197 23.37620 27.78883
## NEFA_sd 0.2211300 0.3132745 0.2202968
## Glucose_sd 24.62831 28.32414 10.32736
## Ketone_sd 0.35162911 0.06555474 0.04030894
## Insulin_sd 8.524563 25.623065 34.903184
## Adiponectin_sd 8.687511 9.339039 9.280763
## Leptin_sd 21.90883 12.33036 10.42255
## RAGE_sd 3.047933 8.984779 3.663867
## CSF_TOT_OD_sd 34.22670 20.82757 10.53853
## CSF_NFL_OD_sd 2.408319 2.037527 2.329000
## CSF_(GCL-IPL)_OD_sd 10.981803 7.072675 4.569331
## CSF_INL_OD_sd 6.800735 5.545268 6.141636
## CSF_ORT_OD_sd 19.34845 13.62373 10.68098
## SI_TOT_OD_sd 16.26743 16.77367 17.22291
## SI_NFL_OD_sd 3.997916 3.028901 1.602555
## SI_(GCL-IPL)_OD_sd 9.130307 9.298094 7.456947
## SI_INL_OD_sd 4.114507 4.679905 4.097301
## SI_ORT_OD_sd 9.408330 7.234178 6.841828
## SO_TOT_OD_sd 12.90978 14.52975 16.01704
## SO_NFL_OD_sd 6.498397 3.725425 2.806918
## SO_(GCL-IPL)_OD_sd 5.761944 5.664215 5.276449
## SO_INL_OD_sd 1.962142 2.022600 2.678478
## SO_ORT_OD_sd 7.311806 7.749878 9.466048
## NI_TOT_OD_sd 19.34673 15.28789 17.23698
## NI_NFL_OD_sd 4.331570 1.928652 1.928652
## NI_(GCL-IPL)_OD_sd 10.258127 8.400487 7.312816
## NI_INL_OD_sd 4.147288 4.944388 5.142662
## NI_ORT_OD_sd 9.843568 8.168676 7.649520
## NO_TOT_OD_sd 17.41455 15.36229 15.78525
## NO_NFL_OD_sd 7.209947 5.940998 4.185111
## NO_(GCL-IPL)_OD_sd 8.053726 6.501165 4.575130
## NO_INL_OD_sd 2.500000 2.015095 2.179449
## NO_ORT_OD_sd 8.262364 9.238834 9.922045
## II_TOT_OD_sd 18.95334 17.46403 15.14376
## II_NFL_OD_sd 3.936475 3.190896 2.015095
## II_(GCL-IPL)_OD_sd 11.074746 8.596599 7.704288
## II_INL_OD_sd 3.872983 4.907477 3.503245
## II_ORT_OD_sd 9.136009 7.716570 8.881373
## IO_TOT_OD_sd 14.38272 15.18647 16.14893
## IO_NFL_OD_sd 7.099296 5.264950 7.751344
## IO_(GCL-IPL)_OD_sd 6.652067 6.359793 5.545268
## IO_INL_OD_sd 1.990603 2.274696 2.958040
## IO_ORT_OD_sd 7.247701 8.225515 8.684713
## TI_TOT_OD_sd 20.64582 16.92139 15.48093
## TI_NFL_OD_sd 2.400521 1.912875 1.083625
## TI_(GCL-IPL)_OD_sd 11.674581 9.354953 9.126634
## TI_INL_OD_sd 3.842200 5.053502 3.553701
## TI_ORT_OD_sd 9.075379 5.915439 7.280110
## TO_TOT_OD_sd 11.99931 14.14294 16.88172
## TO_NFL_OD_sd 2.265686 1.831955 1.544786
## TO_(GCL-IPL)_OD_sd 5.259515 6.200562 6.515134
## TO_INL_OD_sd 2.446085 2.301185 2.839121
## TO_ORT_OD_sd 7.886698 7.075352 8.349832
## CSF_TOT_OS_sd 20.28094 30.51378 20.68303
## CSF_NFL_OS_sd 2.159282 6.612660 1.975225
## CSF_(GCL-IPL)_OS_sd 7.338256 17.849540 7.403930
## CSF_INL_OS_sd 5.019960 7.059809 5.775471
## CSF_ORT_OS_sd 9.871677 11.393778 13.402849
## SI_TOT_OS_sd 12.69383 14.00730 10.16679
## SI_NFL_OS_sd 2.895399 3.369875 2.678478
## SI_(GCL-IPL)_OS_sd 6.886884 7.833011 6.552238
## SI_INL_OS_sd 3.978589 5.477226 3.104249
## SI_ORT_OS_sd 9.589708 9.394954 8.080560
## SO_TOT_OS_sd 11.795479 19.974795 9.802211
## SO_NFL_OS_sd 6.975851 5.264950 3.768892
## SO_(GCL-IPL)_OS_sd 5.218157 6.956031 4.690416
## SO_INL_OS_sd 2.848976 2.810963 2.044949
## SO_ORT_OS_sd 8.119678 9.765431 6.402533
## NI_TOT_OS_sd 14.44472 13.02765 11.25295
## NI_NFL_OS_sd 1.887459 8.099757 2.609714
## NI_(GCL-IPL)_OS_sd 7.379871 9.920899 7.669301
## NI_INL_OS_sd 4.897278 5.783990 2.598076
## NI_ORT_OS_sd 8.715647 9.709024 9.500797
## NO_TOS_OS_sd 14.41238 17.70187 10.28532
## NO_NFL_OS_sd 6.668021 12.932224 5.894913
## NO_GCL-IPL)_OS_sd 4.982887 6.598324 4.399552
## NO_INL_OS_sd 2.657536 4.481443 2.256304
## NO_ORT_OS_sd 25.16843 13.04770 7.00649
## II_TOT_OS_sd 15.39480 14.93902 11.09839
## II_NFL_OS_sd 3.511885 7.673251 2.067058
## II_(GCL-IPL)_OS_sd 7.051891 9.307459 6.562058
## II_INL_OS_sd 4.281258 5.728716 2.895922
## II_ORT_OS_sd 10.467211 9.696532 10.638694
## IO_TOT_OS_sd 14.07939 16.54471 12.19880
## IO_NFL_OS_sd 6.280923 9.159777 4.177864
## IO_(GCL-IPL)_OS_sd 5.402160 7.452882 4.641708
## IO_INL_OS_sd 1.483240 2.796101 2.249579
## IO_ORT_OS_sd 7.899103 10.861386 33.399329
## TI_TOT_OS_sd 13.82374 14.52793 13.54762
## TI_NFL_OS_sd 1.5370426 3.2039275 0.9962049
## TI_(GCL-IPL)_OS_sd 7.576279 8.584694 7.655460
## TI_INL_OS_sd 3.630886 3.904155 3.792936
## TI_ORT_OS_sd 8.215585 6.665151 12.160542
## TO_TOT_OS_sd 13.30100 17.18593 15.30597
## TO_NFL_OS_sd 1.483240 3.107908 1.337116
## TO_(GCL-IPL)_OS_sd 5.588306 7.325237 5.718126
## TO_INL_OS_sd 2.315707 2.005674 2.005674
## TO_OR_OS_sd 8.211526 8.511579 8.674239write.csv(t(summary_group), "summary_group_afterImp.csv", row.names = TRUE, col.names = TRUE)## Warning in write.csv(t(summary_group), "summary_group_afterImp.csv", row.names =
## TRUE, : attempt to set 'col.names' ignoredhistograms of independent variables (age + metabolic measures)
colnames(df.imp)## [1] "IRB" "age" "gender"
## [4] "heart_rate" "MAP" "SBP"
## [7] "BMI" "HbA1c" "BUN"
## [10] "Cr" "BUN/Cr" "HDL"
## [13] "LDL" "Triglycerides" "Cholesterol"
## [16] "NEFA" "Glucose" "Ketone"
## [19] "Insulin" "Adiponectin" "Leptin"
## [22] "RAGE" "CSF_TOT_OD" "CSF_NFL_OD"
## [25] "CSF_(GCL-IPL)_OD" "CSF_INL_OD" "CSF_ORT_OD"
## [28] "SI_TOT_OD" "SI_NFL_OD" "SI_(GCL-IPL)_OD"
## [31] "SI_INL_OD" "SI_ORT_OD" "SO_TOT_OD"
## [34] "SO_NFL_OD" "SO_(GCL-IPL)_OD" "SO_INL_OD"
## [37] "SO_ORT_OD" "NI_TOT_OD" "NI_NFL_OD"
## [40] "NI_(GCL-IPL)_OD" "NI_INL_OD" "NI_ORT_OD"
## [43] "NO_TOT_OD" "NO_NFL_OD" "NO_(GCL-IPL)_OD"
## [46] "NO_INL_OD" "NO_ORT_OD" "II_TOT_OD"
## [49] "II_NFL_OD" "II_(GCL-IPL)_OD" "II_INL_OD"
## [52] "II_ORT_OD" "IO_TOT_OD" "IO_NFL_OD"
## [55] "IO_(GCL-IPL)_OD" "IO_INL_OD" "IO_ORT_OD"
## [58] "TI_TOT_OD" "TI_NFL_OD" "TI_(GCL-IPL)_OD"
## [61] "TI_INL_OD" "TI_ORT_OD" "TO_TOT_OD"
## [64] "TO_NFL_OD" "TO_(GCL-IPL)_OD" "TO_INL_OD"
## [67] "TO_ORT_OD" "CSF_TOT_OS" "CSF_NFL_OS"
## [70] "CSF_(GCL-IPL)_OS" "CSF_INL_OS" "CSF_ORT_OS"
## [73] "SI_TOT_OS" "SI_NFL_OS" "SI_(GCL-IPL)_OS"
## [76] "SI_INL_OS" "SI_ORT_OS" "SO_TOT_OS"
## [79] "SO_NFL_OS" "SO_(GCL-IPL)_OS" "SO_INL_OS"
## [82] "SO_ORT_OS" "NI_TOT_OS" "NI_NFL_OS"
## [85] "NI_(GCL-IPL)_OS" "NI_INL_OS" "NI_ORT_OS"
## [88] "NO_TOS_OS" "NO_NFL_OS" "NO_GCL-IPL)_OS"
## [91] "NO_INL_OS" "NO_ORT_OS" "II_TOT_OS"
## [94] "II_NFL_OS" "II_(GCL-IPL)_OS" "II_INL_OS"
## [97] "II_ORT_OS" "IO_TOT_OS" "IO_NFL_OS"
## [100] "IO_(GCL-IPL)_OS" "IO_INL_OS" "IO_ORT_OS"
## [103] "TI_TOT_OS" "TI_NFL_OS" "TI_(GCL-IPL)_OS"
## [106] "TI_INL_OS" "TI_ORT_OS" "TO_TOT_OS"
## [109] "TO_NFL_OS" "TO_(GCL-IPL)_OS" "TO_INL_OS"
## [112] "TO_OR_OS" "group"indep_list <- c("age", "heart_rate","MAP","SBP","BMI","HbA1c","BUN","Cr","BUN/Cr","HDL","LDL","Triglycerides","Cholesterol","NEFA","Glucose","Ketone","Insulin","Adiponectin","Leptin","RAGE")
for (col in indep_list) {
hist(df.imp[[col]], xlab = paste(col), main = paste("Histogram of", col))
}
ANOVA tests on metabolic measures
# Create an empty list to store the ANOVA test results
anova_results <- list()
p_values <- as.data.frame(cbind(indep_list, rep(NA, length(indep_list))))
colnames(p_values) <- c("Met", "p-value")
# Loop through each column in the vector and perform the ANOVA test
for (col in indep_list) {
my_formula <- formula(paste(col, " ~ group"))
anova_results[[col]] <- aov(my_formula, data = df.imp)
}
# Extract the relevant information from the ANOVA test results
for (col in indep_list) {
anova_result <- anova_results[[col]]
print(paste("Column:", col))
print(summary(anova_result))
p_values[p_values$Met==col, "p-value"] <- summary(anova_result)[[1]][["Pr(>F)"]][[1]]
}## [1] "Column: age"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 35.6 17.79 0.453 0.639
## Residuals 37 1452.0 39.24
## [1] "Column: heart_rate"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 1160 579.8 6.408 0.00408 **
## Residuals 37 3348 90.5
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## [1] "Column: MAP"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 270 134.9 0.879 0.424
## Residuals 37 5677 153.4
## [1] "Column: SBP"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 1476 738.1 1.498 0.237
## Residuals 37 18231 492.7
## [1] "Column: BMI"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 97.9 48.96 1.287 0.288
## Residuals 37 1407.7 38.04
## [1] "Column: HbA1c"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 47.34 23.67 56.32 5.93e-12 ***
## Residuals 37 15.55 0.42
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## [1] "Column: BUN"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 70.5 35.27 1.367 0.267
## Residuals 37 954.4 25.80
## [1] "Column: Cr"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 0.0317 0.01585 0.213 0.809
## Residuals 37 2.7560 0.07449
## [1] "Column: BUN/Cr"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 77.2 38.61 2.38 0.107
## Residuals 37 600.2 16.22
## [1] "Column: HDL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 1181 590.4 1.649 0.206
## Residuals 37 13249 358.1
## [1] "Column: LDL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 5182 2590.9 3.607 0.0371 *
## Residuals 37 26577 718.3
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## [1] "Column: Triglycerides"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 378 188.8 0.104 0.901
## Residuals 37 66866 1807.2
## [1] "Column: Cholesterol"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 10198 5099 4.849 0.0135 *
## Residuals 37 38905 1051
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## [1] "Column: NEFA"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 0.0718 0.03590 0.566 0.573
## Residuals 37 2.3469 0.06343
## [1] "Column: Glucose"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 5848 2923.8 5.665 0.00714 **
## Residuals 37 19096 516.1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## [1] "Column: Ketone"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 0.1554 0.07768 1.497 0.237
## Residuals 37 1.9198 0.05189
## [1] "Column: Insulin"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 1074 536.9 0.915 0.409
## Residuals 37 21713 586.8
## [1] "Column: Adiponectin"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 25.7 12.86 0.157 0.856
## Residuals 37 3038.9 82.13
## [1] "Column: Leptin"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 125 62.26 0.229 0.797
## Residuals 37 10067 272.09
## [1] "Column: RAGE"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 171 85.49 2.692 0.081 .
## Residuals 37 1175 31.76
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1print(p_values)## Met p-value
## 1 age 0.639003947659504
## 2 heart_rate 0.00407830327625814
## 3 MAP 0.423642526330267
## 4 SBP 0.236839174386152
## 5 BMI 0.288160633051946
## 6 HbA1c 5.92743936875186e-12
## 7 BUN 0.267382575964328
## 8 Cr 0.809265749115154
## 9 BUN/Cr 0.106536202914618
## 10 HDL 0.206085579333795
## 11 LDL 0.0370579492213495
## 12 Triglycerides 0.901070924578674
## 13 Cholesterol 0.0134788118727171
## 14 NEFA 0.572660639133777
## 15 Glucose 0.00714012702844985
## 16 Ketone 0.237015529217559
## 17 Insulin 0.40944234717612
## 18 Adiponectin 0.855619965380107
## 19 Leptin 0.796583826653914
## 20 RAGE 0.0810074664704791write.csv(p_values, "ANOVA_met_pValues.csv", row.names = TRUE, col.names = TRUE)## Warning in write.csv(p_values, "ANOVA_met_pValues.csv", row.names = TRUE, :
## attempt to set 'col.names' ignoredImportant predictors: Adiponectin, Leptin, Cholesterol, HbA1c, gender, age
Pairs plot, correlations, heatmap of metabolic measures
pairs(df.imp[met_list])
met_cor <- cor(df.imp[met_list])
corrplot(met_cor, type = "upper", method = "circle")
Selected predictors heatmap
predictors <- c("age","SBP","HbA1c","BUN/Cr","Cholesterol","Ketone","Adiponectin","Leptin","RAGE", "NEFA")
met_cor <- cor(df.imp[predictors])
corrplot(met_cor, type = "upper", method = "circle")
Correlation heatmaps for retinal measures by eye
# right eye
OD_list <- c(23:67)
OD_cor <- cor(df.imp[OD_list])
corrplot(OD_cor, type = "upper", method = "circle", tl.cex = 0.5)
# left eye
OS_list <- c(68:112)
OS_cor <- cor(df.imp[OS_list])
corrplot(OS_cor, type = "upper", method = "circle", tl.cex = 0.5)
# save cleaned dataset
write.table(df.imp, "data_cleaned.txt", sep = "\t", row.names = FALSE, col.names = TRUE)make a long table from a wide one
right_eye_df <- df.imp[,!grepl("_OS",colnames(df.imp),ignore.case = FALSE)]
left_eye_df <- df.imp[,!grepl("_OD",colnames(df.imp),ignore.case = FALSE)]
names(right_eye_df) = gsub(pattern = "_OD",
replacement = "",
x = names(right_eye_df))
names(left_eye_df) = gsub(pattern = "_OS",
replacement = "",
x = names(left_eye_df))
right_eye_df$eye <- "right"
left_eye_df$eye <- "left"
# Fix naming inconsistencies
names(left_eye_df)[names(left_eye_df) == 'TO_OR'] <- 'TO_ORT'
names(left_eye_df)[names(left_eye_df) == 'NO_TOS'] <- 'NO_TOT'
names(left_eye_df)[names(left_eye_df) == 'NO_GCL-IPL)'] <- 'NO_(GCL-IPL)'
df.imp.long <- rbind(right_eye_df,left_eye_df)# save long file
write.table(df.imp.long, "data_cleaned_long.txt", sep = "\t", row.names = FALSE, col.names = TRUE)histograms and descriptive analysis for total retinal thickness in 9 regions (for both eyes)
ret_list <- seq(23, length = 9, by = 5)
for (col in colnames(df.imp.long[ret_list])) {
hist(df.imp.long[[col]], xlab = paste(col), main = paste("Histogram of", col))
}
cols_ind <- c(23:67)
min_data <- apply(df.imp.long[, cols_ind], 2, function(x) min(x, na.rm = TRUE))
max_data <- apply(df.imp.long[, cols_ind], 2, function(x) max(x, na.rm = TRUE))
median_data <- apply(df.imp.long[, cols_ind], 2, function(x) median(x, na.rm = TRUE))
mean_data <- apply(df.imp.long[, cols_ind], 2, function(x) mean(x, na.rm = TRUE))
sd_data <- apply(df.imp.long[, cols_ind], 2, function(x) sd(x, na.rm = TRUE))
n_data <- apply(df.imp.long[, cols_ind], 2, function(x) sum(!is.na(x)))
summary_long <- cbind(min_data, max_data, median_data, mean_data, sd_data, n_data)
colnames(summary_long) <- c("Min", "Max", "Median", "Mean", "SD", "N")
round(summary_long,2)## Min Max Median Mean SD N
## CSF_TOT 231 382 272.5 275.60 25.23 80
## CSF_NFL 8 35 13.0 12.95 3.29 80
## CSF_(GCL-IPL) 23 91 35.0 36.96 10.17 80
## CSF_INL 10 41 22.0 22.15 6.14 80
## CSF_ORT 46 166 115.5 102.80 35.00 80
## SI_TOT 296 370 331.5 333.28 14.60 80
## SI_NFL 18 37 23.0 23.50 3.01 80
## SI_(GCL-IPL) 72 113 88.0 87.70 7.82 80
## SI_INL 33 55 41.0 41.24 4.19 80
## SI_ORT 92 136 116.0 114.97 8.44 80
## SO_TOT 266 340 286.5 290.32 14.18 80
## SO_NFL 29 54 37.0 37.48 5.17 80
## SO_(GCL-IPL) 48 76 57.5 58.38 5.63 80
## SO_INL 25 37 31.0 30.85 2.39 80
## SO_ORT 77 120 98.0 98.14 8.00 80
## NI_TOT 298 380 335.0 335.74 15.44 80
## NI_NFL 16 48 20.0 20.91 4.10 80
## NI_(GCL-IPL) 71 117 89.0 87.90 8.37 80
## NI_INL 33 55 41.0 41.26 4.65 80
## NI_ORT 102 140 122.0 119.85 8.80 80
## NO_TOT 267 345 301.0 303.08 15.23 80
## NO_NFL 35 84 47.0 48.39 7.47 80
## NO_(GCL-IPL) 49 79 60.0 60.62 6.08 80
## NO_INL 27 45 33.0 32.99 2.75 80
## NO_ORT 0 128 96.0 95.10 14.18 80
## II_TOT 291 379 329.5 330.05 15.67 80
## II_NFL 18 48 23.0 23.89 4.07 80
## II_(GCL-IPL) 68 111 85.0 86.00 8.59 80
## II_INL 31 51 41.0 40.94 4.26 80
## II_ORT 94 135 114.5 114.28 9.28 80
## IO_TOT 250 309 279.0 280.88 14.45 80
## IO_NFL 18 63 37.0 37.81 6.63 80
## IO_(GCL-IPL) 45 72 55.5 57.38 5.94 80
## IO_INL 25 37 31.0 30.66 2.29 80
## IO_ORT 74 203 91.0 92.96 14.89 80
## TI_TOT 284 380 321.0 321.99 16.02 80
## TI_NFL 11 27 17.0 17.40 2.02 80
## TI_(GCL-IPL) 66 110 81.5 82.55 9.14 80
## TI_INL 28 49 37.0 37.62 4.03 80
## TI_ORT 97 150 118.5 118.11 8.30 80
## TO_TOT 252 317 274.0 277.21 14.62 80
## TO_NFL 15 29 19.0 19.36 2.11 80
## TO_(GCL-IPL) 51 78 63.0 63.84 5.98 80
## TO_INL 28 38 32.0 32.38 2.32 80
## TO_ORT 75 116 96.0 96.56 7.99 80write.csv(round(summary_long,2), "summary_long.csv", row.names = TRUE, col.names = TRUE)## Warning in write.csv(round(summary_long, 2), "summary_long.csv", row.names =
## TRUE, : attempt to set 'col.names' ignoredcols_ind <- c(23:68)
summary_long_group <- df.imp.long[cols_ind] %>%
group_by(group) %>%
summarize_all(list(median = ~ median(., na.rm = TRUE),
mean = ~ mean(., na.rm = TRUE),
sd = ~ sd(., na.rm = TRUE)))
summary_long_group## # A tibble: 3 × 136
## group CSF_T…¹ CSF_N…² CSF_(…³ CSF_I…⁴ CSF_O…⁵ SI_TO…⁶ SI_NF…⁷ SI_(G…⁸ SI_IN…⁹
## <fct> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 Contr… 278 13.5 37 21.5 120. 330 24 87 41
## 2 DM 278. 13 38 24.5 102. 340. 23 89.5 41
## 3 Predi… 264. 12 33 20 98.5 327 22.5 88 40.5
## # … with 126 more variables: SI_ORT_median <dbl>, SO_TOT_median <dbl>,
## # SO_NFL_median <dbl>, `SO_(GCL-IPL)_median` <dbl>, SO_INL_median <dbl>,
## # SO_ORT_median <dbl>, NI_TOT_median <dbl>, NI_NFL_median <dbl>,
## # `NI_(GCL-IPL)_median` <dbl>, NI_INL_median <dbl>, NI_ORT_median <dbl>,
## # NO_TOT_median <dbl>, NO_NFL_median <dbl>, `NO_(GCL-IPL)_median` <dbl>,
## # NO_INL_median <dbl>, NO_ORT_median <dbl>, II_TOT_median <dbl>,
## # II_NFL_median <dbl>, `II_(GCL-IPL)_median` <dbl>, II_INL_median <dbl>, …write.csv(as.data.frame(t(summary_long_group)), "summary_long_group.csv", row.names = TRUE, col.names = TRUE)## Warning in write.csv(as.data.frame(t(summary_long_group)),
## "summary_long_group.csv", : attempt to set 'col.names' ignoredcols_ind <- seq(23, length = 9, by = 5)
min_data <- apply(df.imp.long[, cols_ind], 2, function(x) min(x, na.rm = TRUE))
max_data <- apply(df.imp.long[, cols_ind], 2, function(x) max(x, na.rm = TRUE))
median_data <- apply(df.imp.long[, cols_ind], 2, function(x) median(x, na.rm = TRUE))
mean_data <- apply(df.imp.long[, cols_ind], 2, function(x) mean(x, na.rm = TRUE))
sd_data <- apply(df.imp.long[, cols_ind], 2, function(x) sd(x, na.rm = TRUE))
n_data <- apply(df.imp.long[, cols_ind], 2, function(x) sum(!is.na(x)))
summary_tot <- cbind(min_data, max_data, median_data, mean_data, sd_data, n_data)
colnames(summary_tot) <- c("Min", "Max", "Median", "Mean", "SD", "N")
round(summary_tot,2)## Min Max Median Mean SD N
## CSF_TOT 231 382 272.5 275.60 25.23 80
## SI_TOT 296 370 331.5 333.28 14.60 80
## SO_TOT 266 340 286.5 290.32 14.18 80
## NI_TOT 298 380 335.0 335.74 15.44 80
## NO_TOT 267 345 301.0 303.08 15.23 80
## II_TOT 291 379 329.5 330.05 15.67 80
## IO_TOT 250 309 279.0 280.88 14.45 80
## TI_TOT 284 380 321.0 321.99 16.02 80
## TO_TOT 252 317 274.0 277.21 14.62 80write.csv(round(summary_tot,2), "summary_tot.csv", row.names = TRUE, col.names = TRUE)## Warning in write.csv(round(summary_tot, 2), "summary_tot.csv", row.names =
## TRUE, : attempt to set 'col.names' ignoredANOVA test for retinal measures
# Create an empty list to store the ANOVA test results
anova_results <- list()
cols_ind <- c(23:67)
ret_list <- colnames(df.imp.long)[cols_ind]
p_values <- as.data.frame(cbind(ret_list, rep(NA, length(ret_list))))
colnames(p_values) <- c("Met", "p-value")
# Loop through each column in the vector and perform the ANOVA test
for (col in ret_list) {
my_formula <- formula(paste("`", col, "`", " ~ group", sep = ""))
anova_results[[col]] <- aov(my_formula, data = df.imp.long)
}
# Extract the relevant information from the ANOVA test results
for (col in ret_list) {
anova_result <- anova_results[[col]]
print(paste("Column:", col))
print(summary(anova_result))
p_values[p_values$Met==col, "p-value"] <- summary(anova_result)[[1]][["Pr(>F)"]][[1]]
}## [1] "Column: CSF_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 4631 2315.7 3.906 0.0242 *
## Residuals 77 45654 592.9
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## [1] "Column: CSF_NFL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 24.1 12.06 1.117 0.333
## Residuals 77 831.7 10.80
## [1] "Column: CSF_(GCL-IPL)"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 465 232.4 2.323 0.105
## Residuals 77 7702 100.0
## [1] "Column: CSF_INL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 204.5 102.25 2.834 0.0649 .
## Residuals 77 2777.7 36.07
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## [1] "Column: CSF_ORT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 659 329.3 0.264 0.769
## Residuals 77 96098 1248.0
## [1] "Column: SI_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 309 154.3 0.719 0.491
## Residuals 77 16533 214.7
## [1] "Column: SI_NFL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 18.7 9.333 1.028 0.363
## Residuals 77 699.3 9.082
## [1] "Column: SI_(GCL-IPL)"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 22 11.00 0.176 0.839
## Residuals 77 4807 62.43
## [1] "Column: SI_INL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 28.3 14.16 0.803 0.452
## Residuals 77 1358.2 17.64
## [1] "Column: SI_ORT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 78 38.84 0.538 0.586
## Residuals 77 5556 72.16
## [1] "Column: SO_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 276 138.0 0.681 0.509
## Residuals 77 15612 202.8
## [1] "Column: SO_NFL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 22.9 11.43 0.421 0.658
## Residuals 77 2091.1 27.16
## [1] "Column: SO_(GCL-IPL)"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 140.6 70.29 2.293 0.108
## Residuals 77 2360.2 30.65
## [1] "Column: SO_INL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 3.3 1.642 0.282 0.755
## Residuals 77 448.9 5.830
## [1] "Column: SO_ORT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 31 15.58 0.239 0.788
## Residuals 77 5022 65.23
## [1] "Column: NI_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 798 398.8 1.703 0.189
## Residuals 77 18028 234.1
## [1] "Column: NI_NFL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 58.2 29.11 1.768 0.178
## Residuals 77 1268.2 16.47
## [1] "Column: NI_(GCL-IPL)"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 16 8.06 0.112 0.894
## Residuals 77 5521 71.70
## [1] "Column: NI_INL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 70.3 35.16 1.656 0.198
## Residuals 77 1635.2 21.24
## [1] "Column: NI_ORT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 75 37.31 0.475 0.623
## Residuals 77 6044 78.49
## [1] "Column: NO_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 255 127.5 0.543 0.583
## Residuals 77 18078 234.8
## [1] "Column: NO_NFL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 41 20.68 0.364 0.696
## Residuals 77 4372 56.77
## [1] "Column: NO_(GCL-IPL)"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 25.5 12.77 0.34 0.713
## Residuals 77 2893.2 37.57
## [1] "Column: NO_INL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 18.8 9.384 1.25 0.292
## Residuals 77 578.2 7.509
## [1] "Column: NO_ORT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 218 109.2 0.537 0.587
## Residuals 77 15657 203.3
## [1] "Column: II_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 472 235.9 0.96 0.387
## Residuals 77 18918 245.7
## [1] "Column: II_NFL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 27.4 13.68 0.821 0.444
## Residuals 77 1282.6 16.66
## [1] "Column: II_(GCL-IPL)"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 31 15.54 0.206 0.814
## Residuals 77 5797 75.28
## [1] "Column: II_INL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 58.4 29.19 1.631 0.202
## Residuals 77 1378.3 17.90
## [1] "Column: II_ORT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 94 46.89 0.538 0.586
## Residuals 77 6714 87.20
## [1] "Column: IO_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 344 172.0 0.82 0.444
## Residuals 77 16141 209.6
## [1] "Column: IO_NFL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 24 11.76 0.262 0.77
## Residuals 77 3451 44.81
## [1] "Column: IO_(GCL-IPL)"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 21.7 10.83 0.302 0.74
## Residuals 77 2763.1 35.88
## [1] "Column: IO_INL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 20.5 10.251 2.007 0.141
## Residuals 77 393.4 5.109
## [1] "Column: IO_ORT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 208 104.1 0.463 0.631
## Residuals 77 17313 224.8
## [1] "Column: TI_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 743 371.3 1.464 0.238
## Residuals 77 19536 253.7
## [1] "Column: TI_NFL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 6.53 3.267 0.799 0.453
## Residuals 77 314.67 4.087
## [1] "Column: TI_(GCL-IPL)"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 109 54.69 0.649 0.525
## Residuals 77 6486 84.24
## [1] "Column: TI_INL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 54.2 27.12 1.703 0.189
## Residuals 77 1226.5 15.93
## [1] "Column: TI_ORT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 37 18.66 0.266 0.767
## Residuals 77 5405 70.19
## [1] "Column: TO_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 385 192.3 0.897 0.412
## Residuals 77 16509 214.4
## [1] "Column: TO_NFL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 8.2 4.098 0.922 0.402
## Residuals 77 342.3 4.445
## [1] "Column: TO_(GCL-IPL)"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 86.2 43.11 1.211 0.303
## Residuals 77 2740.7 35.59
## [1] "Column: TO_INL"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 12.3 6.161 1.145 0.324
## Residuals 77 414.4 5.382
## [1] "Column: TO_ORT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 47 23.32 0.359 0.699
## Residuals 77 4999 64.92print(p_values)## Met p-value
## 1 CSF_TOT 0.0242315563032527
## 2 CSF_NFL 0.332601670228879
## 3 CSF_(GCL-IPL) 0.104813940197745
## 4 CSF_INL 0.0649038143497993
## 5 CSF_ORT 0.768794260212161
## 6 SI_TOT 0.490564491260597
## 7 SI_NFL 0.362703476652037
## 8 SI_(GCL-IPL) 0.838724128959173
## 9 SI_INL 0.451781760035883
## 10 SI_ORT 0.585908445631224
## 11 SO_TOT 0.509257840354474
## 12 SO_NFL 0.657885384999966
## 13 SO_(GCL-IPL) 0.107810485087837
## 14 SO_INL 0.755360207389951
## 15 SO_ORT 0.788137521214331
## 16 NI_TOT 0.188861950544311
## 17 NI_NFL 0.177614191369207
## 18 NI_(GCL-IPL) 0.89384567724524
## 19 NI_INL 0.197682925112661
## 20 NI_ORT 0.623487195680646
## 21 NO_TOT 0.583070104446787
## 22 NO_NFL 0.69595743667068
## 23 NO_(GCL-IPL) 0.712915673197315
## 24 NO_INL 0.292339103279001
## 25 NO_ORT 0.586634156213284
## 26 II_TOT 0.38730970747262
## 27 II_NFL 0.443801666234344
## 28 II_(GCL-IPL) 0.813981134625203
## 29 II_INL 0.202446758788322
## 30 II_ORT 0.586263296048976
## 31 IO_TOT 0.444053166071706
## 32 IO_NFL 0.769866773086835
## 33 IO_(GCL-IPL) 0.740285373751813
## 34 IO_INL 0.141427514840015
## 35 IO_ORT 0.631106787254248
## 36 TI_TOT 0.237765937638517
## 37 TI_NFL 0.453310015375032
## 38 TI_(GCL-IPL) 0.525309660262923
## 39 TI_INL 0.189001278958338
## 40 TI_ORT 0.767251774877591
## 41 TO_TOT 0.412044293626272
## 42 TO_NFL 0.402129568930579
## 43 TO_(GCL-IPL) 0.303449308302904
## 44 TO_INL 0.323647370922986
## 45 TO_ORT 0.699425589751751write.csv(p_values, "ANOVA_ret_pValues.csv", row.names = TRUE, col.names = TRUE)## Warning in write.csv(p_values, "ANOVA_ret_pValues.csv", row.names = TRUE, :
## attempt to set 'col.names' ignoredtotal retinal thickness
# Create an empty list to store the ANOVA test results
anova_results <- list()
cols_ind <- seq(23, length = 9, by = 5)
ret_list <- colnames(df.imp.long)[cols_ind]
p_values <- as.data.frame(cbind(ret_list, rep(NA, length(ret_list))))
colnames(p_values) <- c("Met", "p-value")
# Loop through each column in the vector and perform the ANOVA test
for (col in ret_list) {
my_formula <- formula(paste("`", col, "`", " ~ group", sep = ""))
anova_results[[col]] <- aov(my_formula, data = df.imp.long)
}
# Extract the relevant information from the ANOVA test results
for (col in ret_list) {
anova_result <- anova_results[[col]]
print(paste("Column:", col))
print(summary(anova_result))
p_values[p_values$Met==col, "p-value"] <- summary(anova_result)[[1]][["Pr(>F)"]][[1]]
}## [1] "Column: CSF_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 4631 2315.7 3.906 0.0242 *
## Residuals 77 45654 592.9
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## [1] "Column: SI_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 309 154.3 0.719 0.491
## Residuals 77 16533 214.7
## [1] "Column: SO_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 276 138.0 0.681 0.509
## Residuals 77 15612 202.8
## [1] "Column: NI_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 798 398.8 1.703 0.189
## Residuals 77 18028 234.1
## [1] "Column: NO_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 255 127.5 0.543 0.583
## Residuals 77 18078 234.8
## [1] "Column: II_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 472 235.9 0.96 0.387
## Residuals 77 18918 245.7
## [1] "Column: IO_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 344 172.0 0.82 0.444
## Residuals 77 16141 209.6
## [1] "Column: TI_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 743 371.3 1.464 0.238
## Residuals 77 19536 253.7
## [1] "Column: TO_TOT"
## Df Sum Sq Mean Sq F value Pr(>F)
## group 2 385 192.3 0.897 0.412
## Residuals 77 16509 214.4print(p_values)## Met p-value
## 1 CSF_TOT 0.0242315563032527
## 2 SI_TOT 0.490564491260597
## 3 SO_TOT 0.509257840354474
## 4 NI_TOT 0.188861950544311
## 5 NO_TOT 0.583070104446787
## 6 II_TOT 0.38730970747262
## 7 IO_TOT 0.444053166071706
## 8 TI_TOT 0.237765937638517
## 9 TO_TOT 0.412044293626272write.csv(p_values, "ANOVA_tot_pValues.csv", row.names = TRUE, col.names = TRUE)## Warning in write.csv(p_values, "ANOVA_tot_pValues.csv", row.names = TRUE, :
## attempt to set 'col.names' ignoredTODO: 1. histograms for retinal thickness 2. descriptive by 3 groups? 3. missing values for metabolic measures (REPLACE WITH MEDIAN FOR NOW) 4. A heatmap showing correlations between selected metabolic parameters and retinal thickness values in each of the anatomical locations to select predictors.